Introduction
Anomalous aortic origins of a coronary artery (AAOCA) is a CHD with a prevalence of 0.03–0.9% of the population. Reference Molossi, Doan and Sachdeva1 While most patients with right (R)-AAOCA typically have a benign course, rarely it can lead to sudden cardiac events due to myocardial ischaemia, as it occurs more often in the setting of high-risk left (L)-AAOCA. Reference Doan, Wilkes and O’Neal Reaves2 Some reports have suggested an association between early development of atherosclerosis and AAOCA. Reference Mercado, Coble and Thompson3,Reference Türkay, Gölbasi and Bayezid4 Paediatric patients with CHD are known to have an increased risk of atherosclerotic cardiovascular disease later in life when compared to the general population. Reference Giannakoulas, Dimopoulos and Engel5–Reference Roche and Silversides7 With atherosclerotic cardiovascular disease remaining the leading cause of death in the United States of America, it has been well established that atherosclerotic cardiovascular disease begins in childhood with subclinical atherosclerosis, and hyperlipidaemia correlating to atherosclerotic cardiovascular disease-related adverse events into adulthood. Reference Berenson, Srinivasan, Bao, Newman, Tracy and Wattigney8–Reference Juonala, Viikari and Kähönen12 Coronary artery anomalies represent an understudied yet highly at-risk population for atherosclerotic cardiovascular disease. Reference Zachariah13 To our knowledge, no prior studies have documented lipid profiles in paediatric patients with AAOCA. Given the importance of paediatric intervention for treating hyperlipidaemia in this at-risk population, Reference Magnussen, Koskinen and Juonala9,Reference Juonala, Viikari and Kähönen12,Reference Ference, Yoo and Alesh14–Reference Luirink, Wiegman and Kusters16 we aim to characterise the lipid profiles in paediatric AAOCA patients and explore its relationship to exercise, diagnosis, and demographic factors.
Methods
Study population
Single-center retrospective study of 180 patients ≤ 20 years-old with AAOCA. All patients were prospectively enrolled in our longitudinal database at the Texas Children’s Hospital Coronary Artery Anomalies Program between December 2012 and April 2023. The protocol was approved by the Institutional Review Board, as part of ongoing research study (IRB H-43968, last date approved 09/07/2021). Patients with complex CHD, any aortic valve disease, or hypoplastic coronaries with normal origin were not included in this cohort.
Clinical evaluation and lipid screening
We use an institutionally standardised algorithm (Supplemental Figure S1) for evaluation and management of patients with coronary artery anomalies with best available evidence. Demographics, clinical presentation, electrocardiography, echocardiogram, computed tomography angiography, and advanced imaging with provocative stress were collected for all patients. The standardised approach in our program has been previously published. Reference Molossi, Doan and Sachdeva1,Reference Bonilla-Ramirez, Molossi, Caldarone and Binsalamah17 (Supplemental Figure S1).
Total cholesterol, HDL, LDL, triglycerides, total cholesterol to HDL ratio, and non-HDL cholesterol were obtained retroactively on all patients via electronic health records review at Texas Children’s Hospital and affiliate hospitals and community clinics. Lipid values represent the most recent post-diagnosis, post-surgical (if applicable) lipid panel available in our electronic health records. At least two lipid parameters were needed to be present for data collection. Patients with only one lipid parameter (e.g., triglyceride level only) were not included.
Statistical analysis
Lipid levels were evaluated across demographic (race/ethnicity, sex) as well as clinical characteristics: type of AAOCA (left (L) versus right (R)), presence of ischaemia, exercise level, and surgery status. Exercise levels were classified as recreational or organised/competitive. The normality of data distribution for individual lipid parameters (total cholesterol, triglycerides, LDL, HDL, non-HDL cholesterol, and total cholesterol to HDL ratio) within the cohort data was evaluated using Kolmogorov–Smirnov tests. Accordingly, Mann–Whitney U tests was used for not normally distributed variables (total cholesterol, triglycerides, and total cholesterol to HDL ratio) and two tailed t-tests for approximately normally distributed variables (LDL, HDL, non-HDL cholesterol). The distribution and frequency of lipid levels stratified by sex and race/ethnicity above reference range were determined and presented graphically. Chi-squares were calculated using frequencies above reference range within race/ethnicity groups.
Results
Study population
The median age at diagnosis was 13.7 years (interquartile range 9.7–15.6) with 118 (65.6%) of patients being male (Table 1). The median age of lipid evaluation was 15.6 years (interquartile range 11.6–.66). Race and ethnicities were distributed similarly to our population in Southeast, Texas, with 67 (37%) being of Hispanic origin, 59 (33%) of non-Hispanic (N-H) Black origin, and 48 (27%) of N-H White origin. Most patients (145, 80.5%) had a diagnosis of R-AAOCA versus with L-AAOCA (21, 11.7%). Thirty-nine (22%) patients had demonstrated inducible ischaemia on advanced imaging and 42 (22.8%) underwent coronary surgical repair. All 39 (22%) patients with demonstrated ischaemia were restricted from exercise at initial presentation. Of these, 23 (59%) underwent surgical repair of AAOCA and all but one had exercise restrictions lifted at 3 months postoperatively (Figure 1). Of the remaining 16 (41%) patients with evidence of ischaemia who had not undergone surgical repair, 10 were exercise restricted while awaiting surgery, and 6 were not restricted given shared decision making with the family in the setting of lack of exertional symptoms and conflicting data on provocative testing. At the time of data collection, only 22 (12.2%) patients were restricted from exercise. Of these, 10 with documented ischaemia are awaiting surgery, 1 is awaiting post-operative studies to be completed, and 11 due to high-risk anatomy, exertional symptoms, or pending future preoperative studies. The majority of our patients (121, 67%) exercise recreationally and only 26.7% (48/180) exercising competitively.
Patient distribution with respect to ischaemia, surgery status, and exercise restriction.
Patient demographics
* 1 unknown, 3 Asian, 2 Native American.
† 1 Atretic left main coronary artery, 3 with anomalous circumflex artery.
Lipid profile
Mean or median total cholesterol, LDL, HDL, non-HDL cholesterol, and triglycerides were within reference range (Table 2). When mean or median values of individual lipid parameters were assessed across sex, race/ethnicity, type of AAOCA, presence of ischaemia, surgical status, exercise level, and exercise restriction, only categorisation by race/ethnicity was shown to have statistically significant differences in levels of any individual lipid parameter. Hispanic patients had significantly higher triglycerides (median 99, interquartile range 71–136.5, p = <0.001) and total cholesterol to HDL ratios (median 3.2, interquartile range 2.7–4.5, p = 0.014), versus N-H White N-H Black patients, while total cholesterol approached significance (p = 0.053). Exercise restricted patients had triglycerides above reference range (median = 92 interquartile range 54–112), but were not statistically significant. 18 There was no difference in lipid profiles in those who exercise competitively or in those who were restricted from exercise.
Lipid data (displayed as mean with SD or median with IQR) stratified by patient demographics and compared to accepted reference ranges. Total cholesterol, triglycerides, and total cholesterol to HDL ratio were treated as non-parametric. Reference ranges adapted from Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents 2011
N-H = Non-Hispanic.
Figure 2 demonstrates the distribution of lipid levels by sex and race/ethnicity. The proportion of lipid levels in the abnormal range for all patients in the cohort were total cholesterol (29%, 51/177), HDL (37%, 64/174), triglycerides (44%, 72/165), LDL (16%, 28/170), total cholesterol-HDL ratio (29%, 48/163), and non-HDL cholesterol (28%, 47/165).
Distribution of individual lipid values stratified by sex and race/ethnicity. Red hash line indicates cut-off values of normal and abnormal based off the accepted reference range (adapted from Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents 2011). * p < 0.05.
The majority (71%, 15/21) of Hispanic females and 50% (20/40) of Hispanic males had triglycerides above reference range, while 52% of N-H White females and 8% of N-H Black females had triglycerides above reference range. The frequency of triglyceride levels above the reference range among Hispanic patients was the only lipid parameter found to be statistically significantly elevated within a subgroup defined by race/ethnicity (Figure 2) (χ2(df = 2, n = 160) = 12.36, p = 0.002). Hispanic males also had elevated HDL-total cholesterol ratios (49%, 20/41), versus N-H White (26%) and N-H Black (15%) males, and almost reached significance (p = 0.054). Furthermore, 55% (23/42) of Hispanic males had abnormally low HDL levels versus 24% (6/25) N-H White and 24% (11/45) N-H Black patients (p = 0.09).
Discussion
To our knowledge, this is the first study to document the lipid profile of paediatric patients with AAOCA. While it is reassuring that initial median and mean lipid levels are normal in our population, the frequency of elevated lipid levels in our cohort was noteworthy, particularly in Hispanic patients. It will remain to be seen if the frequencies of abnormal lipid parameters evolve as patients age into adulthood, and if the overall median or mean values may shift as well.
Subclinical atherosclerosis is far more common in childhood, particularly in those with CHD, and the AAOCA population may be at a particular risk given morphologic features of the anomalous coronary vessel, such as acute angle of origin, ostial abnormalities, and intramural course. Reference Molossi, Doan and Sachdeva1,Reference Berenson, Srinivasan, Bao, Newman, Tracy and Wattigney8,Reference Zachariah13,Reference Doan, Zea-Vera and Agrawal19,Reference Suryanarayana, Kollampare and Riaz20
There is an overall paucity of literature surrounding atherosclerosis in the AAOCA population. Some studies have demonstrated an association between early atherosclerotic coronary artery disease in patients with AAOCA. Reference Mercado, Coble and Thompson3,Reference Türkay, Gölbasi and Bayezid4,Reference Shirani and Roberts21–Reference Topaz, Demarchena, Perin, Sommer, Mallon and Chahine23 However, other small studies have not found this association, raising questions of diagnostic bias due to those with AAOCA undergoing coronary angiography earlier than the general adult population. Reference Suryanarayana, Kollampare and Riaz20,Reference Samarendra, Kumari, Hafeez, Vasavada and Sacchi24–Reference Said, de Voogt and Bulut27 Notably, Suryanarayana et al performed the largest study of 148 older adults with anomalous coronary arteries and did not show a greater atherosclerotic burden compared to those with normal coronaries. However, this study was composed of older adults (mean age 69.5 years), with 73% having hypertension, 45% with type 2 diabetes mellitus, and 67% with underlying hyperlipidaemia. Given their age and high risk comorbidities, it raises the question whether these aformentioned risks outweigh those of anomalous coronaries in these older patients and perhaps a younger cohort may indeed delineate AAOCA’s contribution to atherosclorotic risk with greater clarity.
Given elevated cholesterol in youth often precedes vascular and other preclinical atherosclerotic changes, primary care providers following American Academy of Pediatrics lipid screening guidelines remain important. Specifically, focusing on non-HDL cholesterol may be more predictive of persistent dyslipidaemia when compared to LDL, HDL, and total cholesterol alone. Reference Magnussen, Koskinen and Juonala9,Reference Franks, Hanson and Knowler11,Reference Juonala, Viikari and Kähönen12,18,Reference Steinberger, Daniels and Hagberg28 Moreover, given the elevated risk of dyslipidaemia in children with AAOCA, intervening early to reduce lipid levels in childhood offers superior long-term atherosclerotic cardiovascular disease prevention compared to beginning lowering lipids in adulthood. Reference Juonala, Viikari and Kähönen12,Reference Ference, Yoo and Alesh14,Reference Kusters, Avis and de Groot15
Long-term atherosclerotic cardiovascular disease outcomes of AAOCA patients have not been studied. Lui et al. demonstrated that adults with septal cardiac defects were at elevated atherosclerotic cardiovascular disease risk compared to the general population and, furthermore, up to 80% of adults with CHD having at least one atherosclerotic cardiovascular disease risk factor. Reference Giannakoulas, Dimopoulos and Engel5–Reference Roche and Silversides7 Unquestionably, longitudinal studies are needed to evaluate the atherosclerotic cardiovascular disease risk factors and outcomes in AAOCA patients as they age into adulthood and the potential effect of management strategies in its occurrence.
Our study is limited by its retrospective nature and dependence on routine primary care lipid screenings. Our coronary anomaly’s clinic does not routinely test lipid levels. Many primary care providers are practicing outside of the Texas Children’s Hospital network, and thus we are unavailable to obtain these data in some of our patients. Unfortunately, as many do not follow current American Academy of Pediatrics recommended guidelines of pre-pubertal and post-pubertal lipid screens, this likely accounts for why only 180 out of our 573 patients in our cohort have documented lipid profiles. This is consistent with studies identifying deficiencies in lipid screening with roughly a mere 10% of eligible paediatric patients receiving screenings. Reference Sriram, St. Sauver and Jacobson29–Reference Zachariah, McNeal and Copeland31
Future studies to evaluate differences in long-term lipid status in patients with AAOCA and their risk for cardiovascular events are warranted, with particular attention to at-risk racial/ethnic minorities, given frequent exercise restrictions, increasing childhood obesity, and the importance of early intervention for children with dyslipidaemia.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951124035996.
Acknowledgements
We thank all the families AAOCA patients who dedicated time and trust in our team to conduct this study. We also thank the staff at Texas Children’s Hospital in the Lillie Frank Abercrombie Section of Cardiology and especially the coronary anomalies team.
Financial support
None.
Competing interests
The authors have no conflicts to disclose.
Ethical standard
This retrospective chart review study involving human participants was in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The Human Investigation Committee (IRB) of Texas Children’s Hospital and Baylor College of Medicine approved this study.
